]>
jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/generic.py
3 from __future__
import unicode_literals
8 from . common
import InfoExtractor
9 from . youtube
import YoutubeIE
13 compat_urllib_request
,
15 compat_xml_parse_error
,
25 from . brightcove
import BrightcoveIE
26 from . ooyala
import OoyalaIE
27 from . rutv
import RUTVIE
28 from . smotri
import SmotriIE
31 class GenericIE ( InfoExtractor
):
32 IE_DESC
= 'Generic downloader that works on some sites'
37 'url' : 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html' ,
38 'md5' : '85b90ccc9d73b4acd9138d3af4c27f89' ,
40 'id' : '13601338388002' ,
42 'uploader' : 'www.hodiho.fr' ,
43 'title' : 'R\u00e9gis plante sa Jeep' ,
46 # bandcamp page with custom domain
48 'add_ie' : [ 'Bandcamp' ],
49 'url' : 'http://bronyrock.com/track/the-pony-mash' ,
53 'title' : 'The Pony Mash' ,
54 'uploader' : 'M_Pallante' ,
56 'skip' : 'There is a limit of 200 free downloads / month for the test song' ,
58 # embedded brightcove video
59 # it also tests brightcove videos that need to set the 'Referer' in the
62 'add_ie' : [ 'Brightcove' ],
63 'url' : 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/' ,
65 'id' : '2765128793001' ,
67 'title' : 'Le cours de bourse : l’analyse technique' ,
68 'description' : 'md5:7e9ad046e968cb2d1114004aba466fd9' ,
69 'uploader' : 'BFM BUSINESS' ,
72 'skip_download' : True ,
76 # https://github.com/rg3/youtube-dl/issues/2253
77 'url' : 'http://bcove.me/i6nfkrc3' ,
78 'md5' : '0ba9446db037002366bab3b3eb30c88c' ,
80 'id' : '3101154703001' ,
82 'title' : 'Still no power' ,
83 'uploader' : 'thestar.com' ,
84 'description' : 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.' ,
86 'add_ie' : [ 'Brightcove' ],
89 'url' : 'http://www.championat.com/video/football/v/87/87499.html' ,
90 'md5' : 'fb973ecf6e4a78a67453647444222983' ,
92 'id' : '3414141473001' ,
94 'title' : 'Видео. Удаление Дзагоева (ЦСКА)' ,
95 'description' : 'Онлайн-трансляция матча ЦСКА - "Волга"' ,
96 'uploader' : 'Championat' ,
99 # Direct link to a video
101 'url' : 'http://media.w3.org/2010/05/sintel/trailer.mp4' ,
102 'md5' : '67d406c2bcb6af27fa886f31aa934bbe' ,
107 'upload_date' : '20100513' ,
112 'url' : 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219' ,
113 'md5' : '5644c6ca5d5782c1d0d350dad9bd840c' ,
115 'id' : 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ' ,
117 'title' : '2cc213299525360.mov' , # that's what we get
122 'url' : 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE' ,
126 'upload_date' : '20130224' ,
127 'uploader_id' : 'TheVerge' ,
128 'description' : 'Chris Ziegler takes a look at the Alcatel OneTouch Fire and the ZTE Open; two of the first Firefox OS handsets to be officially announced.' ,
129 'uploader' : 'The Verge' ,
130 'title' : 'First Firefox OS phones side-by-side' ,
133 'skip_download' : False ,
138 'url' : 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/' ,
142 'title' : 'Tested: Grinding Coffee at 2000 Frames Per Second' ,
143 'upload_date' : '20140225' ,
144 'description' : 'md5:06a40fbf30b220468f1e0957c0f558ff' ,
145 'uploader' : 'Tested' ,
146 'uploader_id' : 'testedcom' ,
148 # No need to test YoutubeIE here
150 'skip_download' : True ,
155 'url' : 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns' ,
156 'md5' : '7cf780be104d40fea7bae52eed4a470e' ,
160 'title' : 'Between Two Ferns with Zach Galifianakis: President Barack Obama' ,
161 'description' : 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.' ,
166 'url' : 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html' ,
170 'title' : 'Охотское море стало целиком российским' ,
171 'description' : 'md5:5ed62483b14663e2a95ebbe115eb8f43' ,
175 'skip_download' : True ,
180 'url' : 'http://en.support.wordpress.com/videos/ted-talks/' ,
181 'md5' : 'deeeabcc1085eb2ba205474e7235a3d5' ,
185 'title' : 'My web playroom' ,
186 'uploader' : 'Ze Frank' ,
187 'description' : 'md5:ddb2a40ecd6b6a147e400e535874947b' ,
190 # Embeded Ustream video
192 'url' : 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm' ,
193 'md5' : '27b99cdb639c9b12a79bca876a073417' ,
197 'uploader' : 'AU SPA: The NSA and Privacy' ,
198 'title' : 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
201 # nowvideo embed hidden behind percent encoding
203 'url' : 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/' ,
204 'md5' : '2baf4ddd70f697d94b1c18cf796d5107' ,
206 'id' : '06e53103ca9aa' ,
208 'title' : 'Macross Episode 001 Watch Macross Episode 001 onl' ,
209 'description' : 'No description' ,
214 'url' : 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html' ,
215 'md5' : '7653032cbb25bf6c80d80f217055fa43' ,
217 'id' : '048195-004_PLUS7-F' ,
220 'description' : 'md5:d5fdf32ef6613cdbfd516ae658abf168' ,
221 'upload_date' : '20140320' ,
224 'skip_download' : 'Requires rtmpdump'
229 'url' : 'http://rbctv.rbc.ru/archive/news/562949990879132.shtml' ,
230 'md5' : 'ec40048448e9284c9a1de77bb188108b' ,
232 'id' : 'v27008541fad' ,
234 'title' : 'Крым и Севастополь вошли в состав России' ,
235 'description' : 'md5:fae01b61f68984c7bd2fa741e11c3175' ,
237 'upload_date' : '20140318' ,
238 'uploader' : 'rbctv_2012_4' ,
239 'uploader_id' : 'rbctv_2012_4' ,
244 'url' : 'http://www.wired.com/2014/04/honda-asimo/' ,
245 'md5' : 'ba0dfe966fa007657bd1443ee672db0f' ,
247 'id' : '53501be369702d3275860000' ,
249 'title' : 'Honda’s New Asimo Robot Is More Human Than Ever' ,
254 def report_download_webpage ( self
, video_id
):
255 """Report webpage download."""
256 if not self
._ downloader
. params
. get ( 'test' , False ):
257 self
._ downloader
. report_warning ( 'Falling back on generic information extractor.' )
258 super ( GenericIE
, self
). report_download_webpage ( video_id
)
260 def report_following_redirect ( self
, new_url
):
261 """Report information extraction."""
262 self
._ downloader
. to_screen ( '[redirect] Following redirect to %s ' % new_url
)
264 def _send_head ( self
, url
):
265 """Check if it is a redirect, like url shorteners, in case return the new url."""
267 class HEADRedirectHandler ( compat_urllib_request
. HTTPRedirectHandler
):
269 Subclass the HTTPRedirectHandler to make it use our
270 HEADRequest also on the redirected URL
272 def redirect_request ( self
, req
, fp
, code
, msg
, headers
, newurl
):
273 if code
in ( 301 , 302 , 303 , 307 ):
274 newurl
= newurl
. replace ( ' ' , '%20' )
275 newheaders
= dict (( k
, v
) for k
, v
in req
. headers
. items ()
276 if k
. lower () not in ( "content-length" , "content-type" ))
278 # This function was deprecated in python 3.3 and removed in 3.4
279 origin_req_host
= req
. get_origin_req_host ()
280 except AttributeError :
281 origin_req_host
= req
. origin_req_host
282 return HEADRequest ( newurl
,
284 origin_req_host
= origin_req_host
,
287 raise compat_urllib_error
. HTTPError ( req
. get_full_url (), code
, msg
, headers
, fp
)
289 class HTTPMethodFallback ( compat_urllib_request
. BaseHandler
):
291 Fallback to GET if HEAD is not allowed (405 HTTP error)
293 def http_error_405 ( self
, req
, fp
, code
, msg
, headers
):
297 newheaders
= dict (( k
, v
) for k
, v
in req
. headers
. items ()
298 if k
. lower () not in ( "content-length" , "content-type" ))
299 return self
. parent
. open ( compat_urllib_request
. Request ( req
. get_full_url (),
301 origin_req_host
= req
. get_origin_req_host (),
305 opener
= compat_urllib_request
. OpenerDirector ()
306 for handler
in [ compat_urllib_request
. HTTPHandler
, compat_urllib_request
. HTTPDefaultErrorHandler
,
307 HTTPMethodFallback
, HEADRedirectHandler
,
308 compat_urllib_request
. HTTPErrorProcessor
, compat_urllib_request
. HTTPSHandler
]:
309 opener
. add_handler ( handler ())
311 response
= opener
. open ( HEADRequest ( url
))
313 raise ExtractorError ( 'Invalid URL protocol' )
316 def _extract_rss ( self
, url
, video_id
, doc
):
317 playlist_title
= doc
. find ( './channel/title' ). text
318 playlist_desc_el
= doc
. find ( './channel/description' )
319 playlist_desc
= None if playlist_desc_el
is None else playlist_desc_el
. text
323 'url' : e
. find ( 'link' ). text
,
324 'title' : e
. find ( 'title' ). text
,
325 } for e
in doc
. findall ( './channel/item' )]
330 'title' : playlist_title
,
331 'description' : playlist_desc
,
335 def _real_extract ( self
, url
):
336 parsed_url
= compat_urlparse
. urlparse ( url
)
337 if not parsed_url
. scheme
:
338 default_search
= self
._ downloader
. params
. get ( 'default_search' )
339 if default_search
is None :
340 default_search
= 'auto_warning'
342 if default_search
in ( 'auto' , 'auto_warning' ):
344 self
._ downloader
. report_warning ( 'The url doesn \' t specify the protocol, trying with http' )
345 return self
. url_result ( 'http://' + url
)
347 if default_search
== 'auto_warning' :
348 self
._ downloader
. report_warning (
349 'Falling back to youtube search for %s . Set --default-search to "auto" to suppress this warning.' % url
)
350 return self
. url_result ( 'ytsearch:' + url
)
352 assert ':' in default_search
353 return self
. url_result ( default_search
+ url
)
354 video_id
= os
. path
. splitext ( url
. rstrip ( '/' ). split ( '/' )[- 1 ])[ 0 ]
356 self
. to_screen ( ' %s : Requesting header' % video_id
)
359 response
= self
._ send
_ head
( url
)
362 new_url
= response
. geturl ()
364 self
. report_following_redirect ( new_url
)
365 return self
. url_result ( new_url
)
367 # Check for direct link to a video
368 content_type
= response
. headers
. get ( 'Content-Type' , '' )
369 m
= re
. match ( r
'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$' , content_type
)
371 upload_date
= response
. headers
. get ( 'Last-Modified' )
373 upload_date
= unified_strdate ( upload_date
)
376 'title' : os
. path
. splitext ( url_basename ( url
))[ 0 ],
378 'format_id' : m
. group ( 'format_id' ),
380 'vcodec' : 'none' if m
. group ( 'type' ) == 'audio' else None
382 'upload_date' : upload_date
,
385 except compat_urllib_error
. HTTPError
:
386 # This may be a stupid server that doesn't like HEAD, our UA, or so
390 webpage
= self
._ download
_ webpage
( url
, video_id
)
392 # since this is the last-resort InfoExtractor, if
393 # this error is thrown, it'll be thrown here
394 raise ExtractorError ( 'Failed to download URL: %s ' % url
)
396 self
. report_extraction ( video_id
)
400 doc
= parse_xml ( webpage
)
402 return self
._ extract
_ rss
( url
, video_id
, doc
)
403 except compat_xml_parse_error
:
406 # Sometimes embedded video player is hidden behind percent encoding
407 # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
408 # Unescaping the whole page allows to handle those cases in a generic way
409 webpage
= compat_urllib_parse
. unquote ( webpage
)
411 # it's tempting to parse this further, but you would
412 # have to take into account all the variations like
413 # Video Title - Site Name
414 # Site Name | Video Title
415 # Video Title - Tagline | Site Name
416 # and so on and so forth; it's just not practical
417 video_title
= self
._ html
_ search
_ regex
(
418 r
'(?s)<title>(.*?)</title>' , webpage
, 'video title' ,
421 # video uploader is domain name
422 video_uploader
= self
._ search
_ regex
(
423 r
'^(?:https?://)?([^/]*)/.*' , url
, 'video uploader' )
425 # Look for BrightCove:
426 bc_urls
= BrightcoveIE
._ extract
_ brightcove
_u rls
( webpage
)
428 self
. to_screen ( 'Brightcove video detected.' )
431 'url' : smuggle_url ( bc_url
, {'Referer': url}
),
432 'ie_key' : 'Brightcove'
433 } for bc_url
in bc_urls
]
437 'title' : video_title
,
442 # Look for embedded (iframe) Vimeo player
444 r
'<iframe[^>]+?src=(["\' ])( ?P
< url
>( ?
: https?
:) ?
// player\
. vimeo\
. com
/ video
/.+ ?
) \
1 ', webpage)
446 player_url = unescapeHTML(mobj.group(' url
'))
447 surl = smuggle_url(player_url, {'Referer': url} )
448 return self.url_result(surl, ' Vimeo
')
450 # Look for embedded (swf embed) Vimeo player
452 r' < embed
[ ^
>]+ ?src
= "(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)" ', webpage)
454 return self.url_result(mobj.group(1), ' Vimeo
')
456 # Look for embedded YouTube player
457 matches = re.findall(r'''(?x)
458 (?:<iframe[^>]+?src=|embedSWF\(\s*)
459 ([" \' ])(?P<url>(?:https?:)?//(?:www\.)?youtube\.com/
463 urlrs = [self.url_result(unescapeHTML(tuppl[1]), ' Youtube
')
464 for tuppl in matches]
465 return self.playlist_result(
466 urlrs, playlist_id=video_id, playlist_title=video_title)
468 # Look for embedded Dailymotion player
469 matches = re.findall(
470 r' < iframe
[ ^
>]+ ?src
=([ " \' ])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
472 urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Dailymotion')
473 for tuppl in matches]
474 return self.playlist_result(
475 urlrs, playlist_id=video_id, playlist_title=video_title)
477 # Look for embedded Wistia player
479 r'<iframe[^>]+?src=([" \' ])( ?P
< url
>( ?
: https?
:) ?
//( ?
: fast\
.) ?wistia\
. net
/ embed
/ iframe
/.+ ?
) \
1 ', webpage)
482 ' _type
': ' url_transparent
',
483 ' url
': unescapeHTML(match.group(' url
')),
485 ' uploader
': video_uploader,
486 ' title
': video_title,
490 # Look for embedded blip.tv player
491 mobj = re.search(r' < meta\s
[ ^
>]* https?
:// api\
. blip\
. tv
/ \w
+/ redirect
/ \w
+/( \d
+) ', webpage)
493 return self.url_result(' http
:// blip
. tv
/ a
/ a
- '+mobj.group(1), ' BlipTV
')
494 mobj = re.search(r' <( ?
: iframe|embed|
object ) \s
[ ^
>]*( https?
://( ?
: \w
+ \
.) ?blip\
. tv
/( ?
: play
/ |api\
. swf
#)[a-zA-Z0-9]+)', webpage)
496 return self
. url_result ( mobj
. group ( 1 ), 'BlipTV' )
498 # Look for embedded condenast player
499 matches
= re
. findall (
500 r
'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")' ,
507 'ie_key' : 'CondeNast' ,
509 } for ma
in matches
],
510 'title' : video_title
,
514 # Look for Bandcamp pages with custom domain
515 mobj
= re
. search ( r
'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"' , webpage
)
517 burl
= unescapeHTML ( mobj
. group ( 1 ))
518 # Don't set the extractor because it can be a track url or an album
519 return self
. url_result ( burl
)
521 # Look for embedded Vevo player
523 r
'<iframe[^>]+?src=(["\' ])( ?P
< url
>( ?
: https?
:) ?
//( ?
: cache\
.) ?vevo\
. com
/.+ ?
) \
1 ', webpage)
525 return self.url_result(mobj.group(' url
'))
527 # Look for Ooyala videos
528 mobj = (re.search(r' player
. ooyala
. com
/[ ^
"?]+ \? [^" ]* ?
( ?
: embedCode|ec
)=( ?P
< ec
>[ ^
"&]+)', webpage) or
529 re.search(r'OO.Player.create\([ \' " ].* ?
[ \' "],\s*[ \' " ]( ?P
< ec
>. {32}
)[ \' "]', webpage))
531 return OoyalaIE._build_url_result(mobj.group('ec'))
533 # Look for Aparat videos
534 mobj = re.search(r'<iframe .*?src=" ( http
:// www\
. aparat\
. com
/ video
/[ ^
"]+)" ', webpage)
536 return self.url_result(mobj.group(1), ' Aparat
')
538 # Look for MPORA videos
539 mobj = re.search(r' < iframe
.* ?src
= "(http://mpora\.(?:com|de)/videos/[^" ]+) "', webpage)
541 return self.url_result(mobj.group(1), 'Mpora')
543 # Look for embedded NovaMov-based player
545 r'''(?x)<iframe[^>]+?src=([" \' ])
546 ( ?P
< url
> http
://( ?
:( ?
: embed|www
) \
.) ?
548 nowvideo\
.( ?
: ch|sx|eu|at|ag|co
) |
549 videoweed\
.( ?
: es|com
) |
550 movshare\
.( ?
: net|sx|ag
) |
551 divxstage\
.( ?
: eu|net|ch|co|at|ag
))
552 / embed\
. php
.+ ?
) \
1 ''', webpage)
554 return self.url_result(mobj.group('url'))
556 # Look for embedded Facebook player
558 r'<iframe[^>]+?src=([" \' ])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
560 return self.url_result(mobj.group('url'), 'Facebook')
562 # Look for embedded VK player
563 mobj = re.search(r'<iframe[^>]+?src=([" \' ])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
565 return self.url_result(mobj.group('url'), 'VK')
567 # Look for embedded Huffington Post player
569 r'<iframe[^>]+?src=([" \' ])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
571 return self.url_result(mobj.group('url'), 'HuffPost')
574 mobj = re.search(r'class=[" \' ]embedly-card[" \' ][^>]href=[" \' ](?P<url>[^" \' ]+)', webpage)
576 return self.url_result(mobj.group('url'))
577 mobj = re.search(r'class=[" \' ]embedly-embed[" \' ][^>]src=[" \' ][^" \' ]*url=(?P<url>[^&]+)', webpage)
579 return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
581 # Look for funnyordie embed
582 matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
584 urlrs = [self.url_result(unescapeHTML(eurl), 'FunnyOrDie')
586 return self.playlist_result(
587 urlrs, playlist_id=video_id, playlist_title=video_title)
589 # Look for embedded RUTV player
590 rutv_url = RUTVIE._extract_url(webpage)
592 return self.url_result(rutv_url, 'RUTV')
594 # Look for embedded TED player
596 r'<iframe[^>]+?src=([" \' ])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
598 return self.url_result(mobj.group('url'), 'TED')
600 # Look for embedded Ustream videos
602 r'<iframe[^>]+?src=([" \' ])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
604 return self.url_result(mobj.group('url'), 'Ustream')
606 # Look for embedded arte.tv player
608 r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
611 return self.url_result(mobj.group('url'), 'ArteTVEmbed')
613 # Look for embedded smotri.com player
614 smotri_url = SmotriIE._extract_url(webpage)
616 return self.url_result(smotri_url, 'Smotri')
618 # Start with something easy: JW Player in SWFObject
619 mobj = re.search(r'flashvars: [ \' "](?:.*&)?file=(http[^ \' "&]*)', webpage)
621 # Look for gorilla-vid style embedding
622 mobj = re.search(r''' ( ?sx
)
626 jwplayer\s
* \
( \s
*[ "'][^'" ]+[ "']\s*\)\s*\.setup
628 .*?file\s*:\s*[" \' ](.* ?
)[ " \' ]''', webpage)
630 # Broaden the search a little bit
631 mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^ \' " &]*) ', webpage)
633 # Broaden the search a little bit: JWPlayer JS loader
634 mobj = re.search(r' [ ^A
- Za
- z0
- 9 ] ?
file [ " \' ]?:\s*[" \' ]( http ( ?
![ ^
\' "]+\.[0-9]+[ \' " ])[ ^
\' "]+)[" \' ] ', webpage)
637 # Try to find twitter cards info
638 mobj = re.search(r' < meta ( ?
: property |name
)= "twitter:player:stream" ( ?
: content|value
)= "(.+?)" ', webpage)
640 # We look for Open Graph info:
641 # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
642 m_video_type = re.search(r' < meta
.* ?
property = "og:video:type" .* ?content
= "video/(.*?)" ', webpage)
643 # We only look in og:video if the MIME type is a video, don' t
try if it
's a Flash player:
644 if m_video_type is not None:
645 mobj = re.search(r' < meta
.* ?
property = "og:video" .* ?content
= "(.*?)" ', webpage)
648 mobj = re.search(r' < video
[ ^
<]*( ?
:>.* ?
< source
.* ?
) ? src
= "([^" ]+) "', webpage, flags=re.DOTALL)
651 r'(?i)<meta\s+(?=(?:[a-z-]+=" [ ^
"]+" \s
+)* http
- equiv
= "refresh" ) '
652 r' ( ?
:[ a
- z
-]+= "[^" ]+ "\s+)*?content=" [ 0 - 9 ] {,2}
; url
= \' ([ ^
\' ]+) \' "',
655 new_url = mobj.group(1)
656 self.report_following_redirect(new_url)
662 raise ExtractorError('Unsupported URL: %s ' % url)
664 # It's possible that one of the regexes
665 # matched, but returned an empty group:
666 if mobj.group(1) is None:
667 raise ExtractorError('Did not find a valid video URL at %s ' % url)
669 video_url = mobj.group(1)
670 video_url = compat_urlparse.urljoin(url, video_url)
671 video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
673 # Sometimes, jwplayer extraction will result in a YouTube URL
674 if YoutubeIE.suitable(video_url):
675 return self.url_result(video_url, 'Youtube')
677 # here's a fun little line of code for you:
678 video_id = os.path.splitext(video_id)[0]
683 'uploader': video_uploader,
684 'title': video_title,